HotlineLLM

Earn From Compute

Contribute unused CPU and GPU capacity to the HotlineLLM worker network and monetize idle hardware.

Earning Opportunity

Register worker groups with supported models and compute profiles. When your workers match open requests, they process inference and earn based on platform economics—without building your own customer-facing API.

Supported Hardware

  • CPU workstations
  • NVIDIA GPUs
  • Apple Silicon (where supported)
  • Linux servers
  • Private cloud VMs

Installation Process

  1. 1Download the HotlineLLM worker package
  2. 2Configure API credentials and worker group
  3. 3Select models (Ollama or compatible runtime)
  4. 4Start the worker process—it polls for matching tasks
Worker Setup GuideDownload hotlinellm-worker-latest.zip

Worker Rewards

Rewards accrue when your workers complete eligible external tasks. Internal-only workers reduce your organization's AI costs rather than marketplace earnings—choose scope when creating a worker group.

Security Model

Workers authenticate with platform-issued credentials. Tasks are isolated inference jobs—no arbitrary code execution from end users. You define model allowlists and can limit scope to internal workloads only for sensitive environments.

FAQ

How do I earn?

External-scoped workers process marketplace requests. Rewards are tied to completed tasks and platform billing rules.

Is my hardware safe?

Workers only run approved inference workloads. You control which models run and whether you accept internal, external, or hybrid tasks.

What software do I need?

Python worker runtime with Ollama or openai-compatible backends. See the worker setup guide for full requirements.

Create Account & Add Workers